Clean and process TB and population data for Karamoja, Uganda
# Read me
This repository is part of the archive for *Projected impact of climate change on tuberculosis disease burden in Karamoja, Uganda*, part of the CLIMATE project, intended to investigate the effects of climate change on tuberculosis (TB) disease burden, in sub-Saharan East Africa. Specifically, this repository contains the scripts to analyse data from Karamoja, Uganda.
Corresponding author: Fayette Klaassen (owner of this repostory)
## Overview
This repository relies on a non-published data folder `../01-data/`, where the raw, cleaned and processed data are stored. The code in `R/` is used to clean and process the data and create various views and slices of consistent timeseries of TB notification data between 2007 and 2024 in Karamoja, Uganda. Additionally, it creates population estimates (growth rate, mortality and birth rates).
The TB notification data has made available by the Ugandan National TB and Leprosy Program (NTLP), and the population data rely on the Ugandan Bureau of Statistics and the World Bank.
This repository is dependent on R (version xxx), uses `rstan` (version 2.32.6), and `ggplot2`, `tidyverse`, `zoo`, and `readxl` packages (a snapshot of all dependent libraries is available through `renv/lockfile`).
## Detailed schematic overview of data
* note, the data will NOT be uploaded on GitHub, the structure of the files/folders described below is for the internal referencing of files.
- [data-raw/] contains the raw data, as received from partners / downloaded. These are not uploaded on github, but archived on our repository, as not all are for public use. There is further hierarchy here:
- tb/
- 20072014/
- 20152019/
- 20202024/
- turkana/
- population/
- [data-clean/] contains the cleaned data, creating .csv files for the files from the raw data
The files here follow the same folder and name structure as the raw data, to keep clear connection which clean data came from which raw data
- [data-products/] contains the final 'to-use' da …